This paper analyzes a month-long trace of 85 billion user requests and 11.9 million cold starts from Huawei's serverless cloud platform. Our analysis spans workloads from five data centers. We focus on cold starts and provide a comprehensive examination of the underlying factors influencing the number and duration of cold starts. These factors include trigger types, request synchronicity, runtime languages, and function resource allocations. We investigate components of cold starts, including pod allocation time, code and dependency deployment time, and scheduling delays, and examine their relationships with runtime languages, trigger types, and resource allocation. We introduce pod utility ratio to measure the pod's useful lifetime relative to its cold start time, giving a more complete picture of cold starts, and see that some pods with long cold start times have longer useful lifetimes. Our findings reveal the complexity and multifaceted origins of the number, duration, and characteristics of cold starts, driven by differences in trigger types, runtime languages, and function resource allocations. For example, cold starts in Region 1 take up to 7 seconds, dominated by dependency deployment time and scheduling. In Region 2, cold starts take up to 3 seconds and are dominated by pod allocation time. Based on this, we identify opportunities to reduce the number and duration of cold starts using strategies for multi-region scheduling. Finally, we suggest directions for future research to address these challenges and enhance the performance of serverless cloud platforms.

It is challenging to scale time series forecasting models such that they forecast accurately for multiple distinct domains and datasets, all with potentially different underlying collection procedures (e.g., sample resolution), patterns (e.g., periodicity), and prediction requirements (e.g., reconstruction vs. forecasting). We call this general task universal forecasting. Existing methods usually assume that input data is regularly sampled, and they forecast to pre-determined horizons, resulting in failure to generalise outside of the scope of their training. We propose the DAM -- a neural model that takes randomly sampled histories and outputs an adjustable basis composition as a continuous function of time for forecasting to non-fixed horizons. It involves three key components: (1) a flexible approach for using randomly sampled histories from a long-tail distribution, that enables an efficient global perspective of the underlying temporal dynamics while retaining focus on the recent history; (2) a transformer backbone that is trained on these actively sampled histories to produce, as representational output, (3) the basis coefficients of a continuous function of time. We show that a single univariate DAM, trained on 25 time series datasets, either outperformed or closely matched existing SoTA models at multivariate long-term forecasting across 18 datasets, including 8 held-out for zero-shot transfer, even though these models were trained to specialise for each dataset-horizon combination. This single DAM excels at zero-shot transfer and very-long-term forecasting, performs well at imputation, is interpretable via basis function composition and attention, can be tuned for different inference-cost requirements, is robust to missing and irregularly sampled data by design.

This paper releases and analyzes two new Huawei cloud serverless traces. The traces span a period of over 7 months with over 1.4 trillion function invocations combined. The first trace is derived from Huawei's internal workloads and contains detailed per-second statistics for 200 functions running across multiple Huawei cloud data centers. The second trace is a representative workload from Huawei's public FaaS platform. This trace contains per-minute arrival rates for over 5000 functions running in a single Huawei data center. We present the internals of a production FaaS platform by characterizing resource consumption, cold-start times, programming languages used, periodicity, per-second versus per-minute burstiness, correlations, and popularity. Our findings show that there is considerable diversity in how serverless functions behave: requests vary by up to 9 orders of magnitude across functions, with some functions executed over 1 billion times per day; scheduling time, execution time and cold-start distributions vary across 2 to 4 orders of magnitude and have very long tails; and function invocation counts demonstrate strong periodicity for many individual functions and on an aggregate level. Our analysis also highlights the need for further research in estimating resource reservations and time-series prediction to account for the huge diversity in how serverless functions behave.

Efficiently allocating incoming jobs to nodes in large-scale clusters can lead to substantial improvements in both cluster utilization and job performance. In order to allocate incoming jobs, cluster schedulers usually rely on a set of scoring functions to rank feasible nodes. Results from individual scoring functions are usually weighted equally, which could lead to suboptimal deployments as the one-size-fits-all solution does not take into account the characteristics of each workload. Tuning the weights of scoring functions, however, requires expert knowledge and is computationally expensive.This paper proposes a reinforcement learning approach for learning the weights in scheduler scoring algorithms with the overall objective of improving the end-to-end performance of jobs for a given cluster. Our approach is based on percentage improvement reward, frame-stacking, and limiting domain information. We propose a percentage improvement reward to address the objective of multi-step parameter tuning. The inclusion of frame-stacking allows for carrying information across an optimization experiment. Limiting domain information prevents overfitting and improves performance in unseen clusters and workloads. The policy is trained on different combinations of workloads and cluster setups. We demonstrate the proposed approach improves performance on average by 33% compared to fixed weights and 12% compared to the best-performing baseline in a lab-based serverless scenario.

In this work we examine a specific case of wireless sensor networks (WSN) we call peer-to-peer WSN where source and destination are both dynamic and each is subject to constraints of low bandwidth, limited energy storage, and limited computational resources. Peer-to-peer WSN require the consideration of data computation time as a limiting constraint on information availability unlike a standard WSN that can rely on the unconstrained sink to perform the necessary computation of the raw sensor data into usable information. To effectively manage and improve upon peer-to-peer WSN routing, and WSN routing in general, we present a deep reinforcement learning algorithm known as distributed cooperative reinforcement for routing (DCRL-R) which uses a neural network and expanded state space parameters to learn routing policies for WSN. DCRL-R also incorporates an increased action space for determining when and where to perform in-network computation of the raw sensor data. We perform tests of DCRL-R on a physical network utilizing measured node state parametric data and show its viability in future WSN applications compared to a baseline routing algorithm using shortest path decisions with no computational offloading.

Over the years, the popularity of iterative data-intensive applications such as machine learning applications has grown immensely. Unlike batch applications, iterative applications such as k-means, regression or classification algorithms require multiple access to the input data to train it sufficiently for convergence. In the context of big data, these applications are executed on distributed computing frameworks such as Apache Spark. These frameworks are simple to deploy and use, however, under the hood they are complex and highly configurable. To perform an exhaustive study of the impact of these ubiquitous parameters on application performance would be cumbersome due to the exponential amount of their combinations.

Selecting a suitable VM instance type for an application can be a difficult task because of the number of options and the variety of application requirements. Recent research takes a data-driven approach to model VM performance, but this requires carefully choosing a small set of relevant benchmarks as input. We propose a semi-supervised matrix-factorization-based latent variable approach to predict the performance of an unknown new application. This method allows to take a large set of benchmarks as input for VM performance modelling, and it uses the model and the performance measure of the new application on some of the target VMs to predict the performance on the rest of all VMs. We ran experiments with 373 micro-benchmarks from stress-ng and 37 AWS EC2 VMs to predict the scores of Geekbench accurately. Our initial results showed that the RMSE and STD of the predicted scores are 6.7 and 4.5 when sampling Geekbench on 5 VMs, and 10.0 and 2.8 when sampling 10.

In recent years, there has been a lot of focus on benchmarking and performance modelling of data-intensive applications to understand and improve the development of big data systems. Several interesting approaches were proposed; however, as of writing this paper and to the best of our knowledge, there are no comprehensive surveys that thoroughly examine the gaps, trends and trajectories of this area. To fill this void, we, therefore, present a review of the state-of-art benchmarking and performance modelling efforts in data-intensive applications. We start by introducing the two most common dataflow patterns used, for each of these patterns, we review their approach to benchmarking, modelling and validation & experimental environments. Furthermore, we construct a taxonomy and classification to provide a deep understanding of the focus areas of this domain and identify the opportunities for further research. We conclude by analysing each research gap and highlighting future trends.

Modern cluster management systems have been evolving to cope with running and managing diverse cloud applications on heterogeneous computing clusters. Consequently, the system behaviours become complex and non-trivial to explain. In this paper we take the recently published Google trace data set version 3 (V3) as a case study to explore various aspects of intercluster differences. We analyse the distribution of underlying physical machines resource, e.g. number and types of machine, and metrics of computational job requests, e.g. job duration, utilisation and Cycles Per Instruction (CPI). We also apply an unsupervised learning algorithm on the metrics to characterise jobs. Our analysis suggests that the composition of the underlying machine resources in different cells can be substantially different, and the cells with similar machine resource structures can utilise resources differently depending on the characteristics of job requests.

This paper analyzes a newly-published trace that covers 8 different Borg [35] clusters for the month of May 2019. The trace enables researchers to explore how scheduling works in large-scale production compute clusters. We highlight how Borg has evolved and perform a longitudinal comparison of the newly-published 2019 trace against the 2011 trace, which has been highly cited within the research community.

The diversity of cloud offerings motivated the proposition of cloud modelling languages (CMLs) to abstract complexities related to selection of cloud services. However, current CMLs lack the support for modelling service level objectives (SLOs) that are required for the customer applications. Consequently, we propose an application- and provider-independent SLO modelling language (SLO-ML) to enable customers to specify the required SLOs. We also sketch the architecture to realise SLO-ML.

Funding: UK EPSRC EP/R010528/1 and IsDB

How can applications be deployed on the cloud to achieve maximum performance? This question is challenging to address with the availability of a wide variety of cloud Virtual Machines (VMs) with different performance capabilities. The research reported in this paper addresses the above question by proposing a six step benchmarking methodology in which a user provides a set of weights that indicate how important memory, local communication, computation and storage related operations are to an application. The user can either provide a set of four abstract weights or eight fine grain weights based on the knowledge of the application. The weights along with benchmarking data collected from the cloud are used to generate a set of two rankings-one based only on the performance of the VMs and the other takes both performance and costs into account. The rankings are validated on three case study applications using two validation techniques. The case studies on a set of experimental VMs highlight that maximum performance can be achieved by the three top ranked VMs and maximum performance in a cost-effective manner is achieved by at least one of the top three ranked VMs produced by the methodology.

Cloud computing is a ubiquitous part of the computing landscape. For many companies today, moving their computing infrastructure to the cloud reduces time to deployment and saves money. Spot Instances, a subset of Amazon's cloud computing infrastructure (EC2), expands upon this. They allow a user to bid on spare compute capacity in EC2 at heavily discounted prices. If other bids for the spare capacity exceeds the user's own, the user's instance is terminated. In this paper, we conduct one of the first detailed analyses of how location affects the overall cost of deployment of a Spot Instance. We analyse pricing data across all available AWS regions for 60 days for a variety of Spot Instances. We relate the pricing data we find to the overall AWS region and examine any patterns we see across the week. We find that location plays a critical role in Spot Instance pricing and that pricing differs, sometimes markedly, from region to region. We conclude by showing that it is indeed possible to run workloads on Spot Instances with low risk of termination and a low overall cost.

The recent boom of big data, coupled with the challenges of its processing and storage gave rise to the development of distributed data processing and storage paradigms like MapReduce, Spark, and NoSQL databases. With the advent of cloud computing, processing and storing such massive datasets on clusters of machines is now feasible with ease. However, there are limited tools and approaches, which users can rely on to gauge and comprehend the performance of their big data applications deployed locally on clusters, or in the cloud. Researchers have started exploring this area by providing benchmarking suites suitable for big data applications. However, many of these tools are fragmented, complex to deploy and manage, and do not provide transparency with respect to the monetary cost of benchmarking an application. In this paper, we present Plug And Play Bench (PAPB1): an infrastructure aware abstraction built to integrate and simplify the deployment of big data benchmarking tools on clusters of machines. PAPB automates the tedious process of installing, configuring and executing common big data benchmark workloads by containerising the tools and settings based on the underlying cluster deployment framework. Our proof of concept implementation utilises HiBench as the benchmark suite, HDP as the cluster deployment framework and Azure as the cloud platform. The paper further illustrates the inclusion of cost metrics based on the underlying Microsoft Azure cloud platform.

Cloud computing providers such as Amazon and Google have recently begun offering container-instances, which provide an efficient route to application deployment within a lightweight, isolated and well-defined execution environment. Cloud providers currently offer Container Service Platforms (CSPs), which orchestrate containerised applications. Existing CSP frameworks do not offer any form of intelligent resource scheduling: applications are usually scheduled individually, rather than taking a holistic view of all registered applications and available resources in the cloud. This can result in increased execution times for applications, resource wastage through underutilised container-instances, and a reduction in the number of applications that can be deployed, given the available resources. The research presented in this paper aims to extend existing systems by adding a cloud-based Container Management Service (CMS) framework that offers increased deployment density, scalability and resource efficiency. CMS provides additional functionalities for orchestrating containerised applications by joint optimisation of sets of containerised applications, and resource pool in multiple (geographical distributed) cloud regions. We evaluated CMS on a cloud-based CSP i.e., Amazon EC2 Container Management Service (ECS) and conducted extensive experiments using sets of CPU and Memory intensive containerised applications against the direct deployment strategy of Amazon ECS. The results show that CMS achieves up to 25% higher cluster utilisation, and up to 70% reduction in execution times.

This paper presents MAMoC, a framework which brings together a diverse range of infrastructure types including mobile devices, cloudlets, and remote cloud resources under one unified API. MAMoC allows mobile applications to leverage the power of multiple offloading destinations. MAMoC's intelligent offloading decision engine adapts to the contextual changes in this heterogeneous environment, in order to reduce the overall runtime for both single-site and multi-site offloading scenarios. MAMoC is evaluated through a set of offloading experiments, which evaluate the performance of our offloading decision engine. The results show that offloading computation using our framework can reduce the overall task completion time for both single-site and multi-site offloading scenarios.

Existing benchmarking methods are time consuming processes as they typically benchmark the entire Virtual Machine (VM) in order to generate accurate performance data, making them less suitable for real-time analytics. The research in this paper is aimed to surmount the above challenge by presenting DocLite - Docker Container-based Lightweight benchmarking tool. DocLite explores lightweight cloud benchmarking methods for rapidly executing benchmarks in near real-time. DocLite is built on the Docker container technology, which allows a user-defined memory size and number of CPU cores of the VM to be benchmarked. The tool incorporates two benchmarking methods - the first referred to as the native method employs containers to benchmark a small portion of the VM and generate performance ranks, and the second uses historic benchmark data along with the native method as a hybrid to generate VM ranks. The proposed methods are evaluated on three use-cases and are observed to be up to 91 times faster than benchmarking the entire VM. In both methods, small containers provide the same quality of rankings as a large container. The native method generates ranks with over 90% and 86% accuracy for sequential and parallel execution of an application compared against benchmarking the whole VM. The hybrid method did not improve the quality of the rankings significantly.

It is challenging to execute an application in a heterogeneous cloud cluster, which consists of multiple types of virtual machines with different performance capabilities and prices. This paper aims to mitigate this challenge by proposing a scheduling mechanism to optimise the execution of Bag-of-Task jobs on a heterogeneous cloud cluster. The proposed scheduler considers two approaches to select suitable cloud resources for executing a user application while satisfying pre-defined Service Level Objectives (SLOs) both in terms of execution deadline and minimising monetary cost. Additionally, a mechanism for dynamic re-assignment of jobs during execution is presented to resolve potential violation of SLOs. Experimental studies are performed both in simulation and on a public cloud using real-world applications. The results highlight that our scheduling approaches result in cost saving of up to 31% in comparison to naive approaches that only employ a single type of virtual machine in a homogeneous cluster. Dynamic reassignment completely prevents deadline violation in the best-case and reduces deadline violations by 95% in the worst-case scenario.

With the availability of a wide range of cloud Virtual Machines (VMs) it is difficult to determine which VMs can maximise the performance of an application. Benchmarking is commonly used to this end for capturing the performance of VMs. Most cloud benchmarking techniques are typically heavyweight - time consuming processes which have to benchmark the entire VM in order to obtain accurate benchmark data. Such benchmarks cannot be used in real-time on the cloud and incur extra costs even before an application is deployed. In this paper, we present lightweight cloud benchmarking techniques that execute quickly and can be used in near real-time on the cloud. The exploration of lightweight benchmarking techniques are facilitated by the development of DocLite - Docker Container-based Lightweight Benchmarking. DocLite is built on the Docker container technology which allows a user-defined portion (such as memory size and the number of CPU cores) of the VM to be benchmarked. DocLite operates in two modes, in the first mode, containers are used to benchmark a small portion of the VM to generate performance ranks. In the second mode, historic benchmark data is used along with the first mode as a hybrid to generate VM ranks. The generated ranks are evaluated against three scientific high-performance computing applications. The proposed techniques are up to 91 times faster than a heavyweight technique which benchmarks the entire VM. It is observed that the first mode can generate ranks with over 90% and 86% accuracy for sequential and parallel execution of an application. The hybrid mode improves the correlation slightly but the first mode is sufficient for benchmarking cloud VMs.

Summary Orchestrating workflows needed for modern scientific data analysis presents a significant research challenge: they are typically executed in a centralized manner such that all data pass through a single compute server known as the engine, which causes unnecessary network traffic that leads to a performance bottleneck. This paper presents a scalable decentralized orchestration system that relies on a functional, high‐level data coordination language for executing workflows. This system consists of distributed execution engines, each of which is responsible for executing part of the overall workflow. It exploits parallelism in the workflow by partitioning it into smaller sub‐workflows and determines the most appropriate engines to execute them using network resource monitoring and placement analysis. This permits the computation logic of the workflow to be moved towards the services providing the data, which improves the overall execution time. The system supports data‐driven execution that allows each sub‐workflow to be executed as soon as the data needed for its execution become available from other sources. Therefore, a scheduling mechanism is not required to manage the order in which the sub‐workflows are orchestrated. This paper provides an evaluation of the proposed system, which demonstrates that decentralized orchestration provides scalability over centralized orchestration. Copyright © 2015 John Wiley & Sons, Ltd.

A Cloud Services Brokerage (CSB) acts as an intermediary between cloud service providers (e.g., Amazon and Google) and cloud service end users, providing a number of value adding services. CSBs as a research topic are in there infancy. The goal of this paper is to provide a concise survey of existing CSB technologies in a variety of areas and highlight a roadmap, which details five future opportunities for research.

This paper presents the first complete, integrated and end-to-end solution for ad hoc cloud computing environments. Ad hoc clouds harvest resources from existing sporadically available, non-exclusive (i.e. Primarily used for some other purpose) and unreliable infrastructures. In this paper we discuss the problems ad hoc cloud computing solves and outline our architecture which is based on BOINC.

Optimising the execution of Bag-of-Tasks (BoT) applications on the cloud is a hard problem due to the trade-offs between performance and monetary cost. The problem can be further complicated when multiple BoT applications need to be executed. In this paper, we propose and implement a heuristic algorithm that schedules tasks of multiple applications onto different cloud virtual machines in order to maximise performance while satisfying a given budget constraint. Current approaches are limited in task scheduling since they place a limit on the number of cloud resources that can be employed by the applications. However, in the proposed algorithm there are no such limits, and in comparison with other approaches, the algorithm on average achieves an improved performance of 10%. The experimental results also highlight that the algorithm yields consistent performance even with low budget constraints which cannot be achieved by competing approaches.

This research is supported by the EPSRC grant ‘Working Together: Constraint Programming and Cloud Computing’ (EP/K015745/1), a Royal Society Industry Fellowship, an Impact Acceleration Account Grant (IAA) and an Amazon Web Services (AWS) Education Research Grant.

Gravitational microlensing exploits a transient phenomenon where an observed star is brightened due to deflection of its light by the gravity of an intervening foreground star. It is conjectured that this technique can be used to measure the abundance of planets throughout the Milky Way. In order to undertake efficient gravitational microlensing an observation schedule must be constructed such that various targets are observed while undergoing a microlensing event. In this paper, we propose a cloud-based e-Infrastructure that currently supports four methods to compute candidate schedules via the application of local search and probabilistic meta-heuristics. We then validate the feasibility of the e-Infrastructure by evaluating the methods on historic data. The experiments demonstrate that the use of on-demand cloud resources for the e-Infrastructure can allow better schedules to be found more rapidly.

Scheduling Bag-of-Tasks (BoT) applications on the cloud can be more challenging than grid and cluster environments. This is because a user may have a budgetary constraint or a deadline for executing the BoT application in order to keep the overall execution costs low. The research in this paper is motivated to investigate task scheduling on the cloud, given two hard constraints based on a user-defined budget and a deadline. A heuristic algorithm is proposed and implemented to satisfy the hard constraints for executing the BoT application in a cost effective manner. The proposed algorithm is evaluated using four scenarios that are based on the trade-off between performance and the cost of using different cloud resource types. The experimental evaluation confirms the feasibility of the algorithm in satisfying the constraints. The key observation is that multiple resource types can be a better alternative to using a single type of resource.

Summary Rising complexity within multi‐tier computing architectures remains an open problem. As complexity increases, so do the costs associated with operating and maintaining systems within these environments. One approach for addressing these problems is to buildself‐healingsystems (i.e. frameworks) that can autonomously detect and recover from faulty states. Self‐healing systems often combine machine learning techniques with closed control loops to reduce the number of situations requiring human intervention. This is particularly useful in situations where human involvement is both costly to develop, and a source of potential faults. Therefore, a survey of self‐healing frameworks and methodologies in multi‐tier architectures is provided to the reader. Uniquely, this study combines an overview of the state of the art with a comparative analysis of the computing environment, degree of behavioural autonomy, and organisational requirements of these approaches. Highlighting these aspects provides for an understanding of the different situational benefits of these self‐healing systems. We conclude with a discussion of potential and current research directions within this field. Copyright © 2014 John Wiley & Sons, Ltd.

Large-scale ad hoc analytics of genomic data is popular using the R-programming language supported by over 700 software packages provided by Bioconductor. More recently, analytical jobs are benefitting from on-demand computing and storage, their scalability and their low maintenance cost, all of which are offered by the cloud. While biologists and bioinformaticists can take an analytical job and execute it on their personal workstations, it remains challenging to seamlessly execute the job on the cloud infrastructure without extensive knowledge of the cloud dashboard. How analytical jobs can not only with minimum effort be executed on the cloud, but also how both the resources and data required by the job can be managed is explored in this paper. An open-source light-weight framework for executing R-scripts using Bioconductor packages, referred to as `RBioCloud', is designed and developed. RBioCloud offers a set of simple command-line tools for managing the cloud resources, the data and the execution of the job. Three biological test cases validate the feasibility of RBioCloud. The framework is available from http://www.rbiocloud.com.

This paper argues that there are three fundamental challenges that need to be overcome in order to foster the adoption of big data technologies in non-computer science related disciplines: addressing issues of accessibility of such technologies for non-computer scientists, supporting the ad hoc exploration of large data sets with minimal effort and the availability of lightweight web-based frameworks for quick and easy analytics. In this paper, we address the above three challenges through the development of `BigExcel', a three tier web-based framework for exploring big data to facilitate the management of user interactions with large data sets, the construction of queries to explore the data set and the management of the infrastructure. The feasibility of BigExcel is demonstrated through two Yahoo Sandbox datasets. The first dataset is the Yahoo Buzz Score data set we use for quantitatively predicting trending technologies and the second is the Yahoo n-gram corpus we use for qualitatively inferring the coverage of important events. A demonstration of the BigExcel framework and source code is available at http://bigdata.cs.st-andrews.ac. uk/projects/bigexcel-exploring-big-data-for-social-sciences/.

When orchestrating Web service workflows, the geographical placement of the orchestration engine (s) can greatly affect workflow performance. Data may have to be transferred across long geographical distances, which in turn increases execution time and degrades the overall performance of a workflow. In this paper, we present a framework that, given a DAG-based workflow specification, computes the optimal Amazon EC2 cloud regions to deploy the orchestration engines and execute a workflow. The framework incorporates a constraint model that solves the workflow deployment problem, which is generated using an automated constraint modelling system. The feasibility of the framework is evaluated by executing different sample workflows representative of scientific workloads. The experimental results indicate that the framework reduces the workflow execution time and provides a speed up of 1.3x-2.5x over centralised approaches.

Bag of Distributed Tasks (BoDT) can benefit from decentralised execution on the Cloud. However, there is a trade-off between the performance that can be achieved by employing a large number of Cloud VMs for the tasks and the monetary constraints that are often placed by a user. The research reported in this paper is motivated towards investigating this trade-off so that an optimal plan for deploying BoDT applications on the cloud can be generated. A heuristic algorithm, which considers the user's preference of performance and cost is proposed and implemented. The feasibility of the algorithm is demonstrated by generating execution plans for a sample application. The key result is that the algorithm generates optimal execution plans for the application over 91% of the time.

How can applications be deployed on the cloud to achieve maximum performance? This question has become significant and challenging with the availability of a wide variety of Virtual Machines (VMs) with different performance capabilities in the cloud. The above question is addressed by proposing a six step benchmarking methodology in which a user provides a set of four weights that indicate how important each of the following groups: memory, processor, computation and storage are to the application that needs to be executed on the cloud. The weights along with cloud benchmarking data are used to generate a ranking of VMs that can maximise performance of the application. The rankings are validated through an empirical analysis using two case study applications, the first is a financial risk application and the second is a molecular dynamics simulation, which are both representative of workloads that can benefit from execution on the cloud. Both case studies validate the feasibility of the methodology and highlight that maximum performance can be achieved on the cloud by selecting the top ranked VMs produced by the methodology.

Analysis of information retrieved from microblog-ging services such as Twitter can provide valuable insight into public sentiment in a geographic region. This insight can be enriched by visualising information in its geographic context. Two underlying approaches for sentiment analysis are dictionary-based and machine learning. The former is popular for public sentiment analysis, and the latter has found limited use for aggregating public sentiment from Twitter data. The research presented in this paper aims to extend the machine learning approach for aggregating public sentiment. To this end, a framework for analysing and visualising public sentiment from a Twitter corpus is developed. A dictionary-based approach and a machine learning approach are implemented within the framework and compared using one UK case study, namely the royal birth of 2013. The case study validates the feasibility of the framework for analysis and rapid visualisation. One observation is that there is good correlation between the results produced by the popular dictionary-based approach and the machine learning approach when large volumes of tweets are analysed. However, for rapid analysis to be possible faster methods need to be developed using big data techniques and parallel methods.

The Berkeley Open Infrastructure for Network Computing (BOINC) is an open source client-server middleware system created to allow projects with large computational requirements, usually set in the scientific domain, to utilize a technically unlimited number of volunteer machines distributed over large physical distances. However various problems exist deploying applications over these heterogeneous machines using BOINC: applications must be ported to each machine architecture type, the project server must be trusted to supply authentic applications, applications that do not regularly checkpoint may lose execution progress upon volunteer machine termination and applications that have dependencies may find it difficult to run under BOINC. To solve such problems we introduce virtual BOINC, or V-BOINC, where virtual machines are used to run computations on volunteer machines. Application developers can then compile their applications on a single architecture, check pointing issues are solved through virtualization API's and many security concerns are addressed via the virtual machine's sandbox environment. In this paper we focus on outlining a unique approach on how virtualization can be introduced into BOINC and demonstrate that V-BOINC offers acceptable computational performance when compared to regular BOINC. Finally we show that applications with dependencies can easily run under V-BOINC in turn increasing the computational potential volunteer computing offers to the general public and project developers.

When orchestrating highly distributed and data-intensive Web service workflows the geographical placement of the orchestration engine can greatly affect the overall performance of a workflow. Orchestration engines are typically run from within an organisations' network, and may have to transfer data across long geographical distances, which in turn increases execution time and degrades the overall performance of a workflow. In this paper we present Cloud Forecast: a Web service framework and analysis tool which given a workflow specification, computes the optimal Amazon EC2 Cloud region to automatically deploy the orchestration engine and execute the workflow. We use geographical distance of the workflow, network latency and HTTP round-trip time between Amazon Cloud regions and the workflow nodes to find a ranking of Cloud regions. This combined set of simple metrics effectively predicts where the workflow orchestration engine should be deployed in order to reduce overall execution time. We evaluate our approach by executing randomly generated data-intensive workflows deployed on the Planet Lab platform in order to rank Amazon EC2 Cloud regions. Our experimental results show that our proposed optimisation strategy, depending on the particular workflow, can speed up execution time on average by 82.25% compared to local execution. We also show that the standard deviation of execution time is reduced by an average of almost 65% using the optimisation strategy.

Server clustering is a common design principle employed by many organisations who require high availability, scalability and easier management of their infrastructure. Servers are typically clustered according to the service they provide whether it be the application(s) installed, the role of the server or server accessibility for example. In order to optimize performance, manage load and maintain availability, servers may migrate from one cluster group to another making it difficult for server monitoring tools to continuously monitor these dynamically changing groups. Server monitoring tools are usually statically configured and with any change of group membership requires manual reconfiguration, an unreasonable task to undertake on large-scale cloud infrastructures. In this paper we present the Cloudlet Control and Management System (C2MS), a system for monitoring and controlling dynamic groups of physical or virtual servers within cloud infrastructures. The C2MS extends Ganglia - an open source scalable system performance monitoring tool - by allowing system administrators to define, monitor and modify server groups without the need for server reconfiguration. In turn administrators can easily monitor group and individual server metrics on large-scale dynamic cloud infrastructures where roles of servers may change frequently. Furthermore, we complement group monitoring with a control element allowing administrator-specified actions to be performed over servers within service groups as well as introduce further customized monitoring metrics. This paper outlines the design, implementation and evaluation of the C2MS.

Monitoring is an essential aspect of maintaining and developing computer systems which increases in difficulty proportional to the size of the system. The need for robust monitoring tools has become more evident with the advent of cloud computing. Infrastructure as a Service (IaaS) clouds allow end users to deploy vast numbers of virtual machines as part of dynamic and transient architectures. Current monitoring solutions, including many of those in the open-source domain, rely on outdated concepts including manual configuration and centralised data collection and adapt poorly to membership churn. In this paper we propose the development of a cloud monitoring system to provide scalable and robust lookup, data collection and analysis services for large-scale cloud systems. In lieu of centrally managed monitoring we propose a multi-tier architecture using a layered gossip protocol to aggregate monitoring information and facilitate lookup, information collection and the identification of redundant capacity. This allows for a resource aware data collection and storage architecture that operates over the system being monitored. This in turn enables monitoring to be done in situ without the need for significant additional infrastructure to facilitate monitoring services. We evaluate this approach against alternative monitoring paradigms and demonstrate how our solution is well adapted to usage in a cloud-computing context.

Service-oriented workflows are typically executed using a centralised orchestration approach that presents significant scalability challenges. These challenges include the consumption of network bandwidth, degradation of performance, and single-points of failure. We provide a decentralised orchestration architecture that attempts to address these challenges. Our architecture adopts a design model that permits the computation to be moved "closer" to services in a workflow. This is achieved by partitioning workflows specified using our simple dataflow language into smaller fragments, which may be sent to remote locations for execution.

Orchestrating centralised service-oriented workflows presents significant scalability challenges that include: the consumption of network bandwidth, degradation of performance, and single points of failure. This paper presents a high-level dataflow specification language that attempts to address these scalability challenges. This language provides simple abstractions for orchestrating large-scale web service workflows, and separates between the workflow logic and its execution. It is based on a data-driven model that permits parallelism to improve the workflow performance. We provide a decentralised architecture that allows the computation logic to be moved "closer" to services involved in the workflow. This is achieved through partitioning the workflow specification into smaller fragments that may be sent to remote orchestration services for execution. The orchestration services rely on proxies that exploit connectivity to services in the workflow. These proxies perform service invocations and compositions on behalf of the orchestration services, and carry out data collection, retrieval, and mediation tasks. The evaluation of our architecture implementation concludes that our decentralised approach reduces the execution time of workflows, and scales accordingly with the increasing size of data sets.

As the number of services and the size of data involved in workflows increases, centralized orchestration techniques are reaching the limits of scalability. When relying on web services without third-party data transfer, a standard orchestration model needs to pass all data through a centralized engine, which results in unnecessary data transfer and the engine to become a bottleneck to the execution of a workflow. As a solution, this paper presents and evaluates Circulate, an alternative service-oriented architecture which facilitates an orchestration model of central control in combination with a choreography model of optimized distributed data transport. Extensive performance analysis through the PlanetLab framework is conducted on a web service-based implementation over a range of Internet-scale configurations which mirror scientific workflow environments. Performance analysis concludes that our architecture's optimized model of data transport speeds up the execution time of workflows, consistently outperforms standard orchestration and scales with data and node size. Furthermore, Circulate is a less-intrusive solution as individual services do not have to be reconfigured in order to take part in a workflow.

Cloud computing is an emerging technology that allows users to utilize on-demand computation, storage, data and services from around the world. However, Cloud service providers charge users for these services. Specifically, to access data from their globally distributed storage edge servers, providers charge users depending on the user's location and the amount of data transferred. When deploying data-intensive applications in a Cloud computing environment, optimizing the cost of transferring data to and from these edge servers is a priority, as data play the dominant role in the application's execution. In this paper, we formulate a non-linear programming model to minimize the data retrieval and execution cost of data-intensive workflows in Clouds. Our model retrieves data from Cloud storage resources such that the amount of data transferred is inversely proportional to the communication cost. We take an example of an `intrusion detection' application workflow, where the data logs are made available from globally distributed Cloud storage servers. We construct the application as a workflow and experiment with Cloud based storage and compute resources. We compare the cost of multiple executions of the workflow given by a solution of our non-linear program against that given by Amazon CloudFront's `nearest' single data source selection. Our results show a savings of three-quarters of total cost using our model.

Within the context of the EU design study developmental gene expression map, we identify a set of challenges when facilitating collaborative research on early human embryo development. These challenges bring forth requirements, for which we have identified solutions and technology. We summarise our solutions and demonstrate how they integrate to form an e-infrastructure to support collaborative research in this area of developmental biology.

This paper introduces the multiagent protocols (MAP) Web service choreography language and demonstrates how service choreographies can be specified, verified, and enacted with a comparatively simple process language. MAP is a directly executable specification, services do not have to be preconfigured at design-time. Instead, a choreography, specified in MAP, can be sent dynamically to a group of distributed peers to execute at runtime. Furthermore, MAP is based on a formal foundation, this allows model checking of the choreography definition prior to live distribution and enactment. A motivating scenario, taken from the AstroGrid science use-cases, serves as the focal point for the paper and highlights the benefits of choreography, through data flow optimization and lack of centralized server. The MAP formal syntax and model checking environment are discussed in the context of the motivating scenario, along with MagentA, an implementation of MAP which provides a concrete, and open-source framework for the enactment of distributed choreographies. MAP is evaluated by demonstrating the languages conformance to the service interaction patterns, a collection of 13 recurring workflow patterns.

When orchestrating data-centric workflows as are commonly found in the sciences, centralised servers can become a bottleneck to the performance of a workflow; output from service invocations are normally transferred via a centralised orchestration engine, when they should be passed directly to where they are needed at the next service in the workflow. To address this performance bottleneck, this paper presents a lightweight hybrid workflow architecture and concrete API, based on a centralised control flow, distributed data flow model. Our architecture maintains the robustness and simplicity of centralised orchestration, but facilitates choreography by allowing services to exchange data directly with one another, reducing data that needs to be transferred through a centralised server. Furthermore our architecture is standards compliment, flexible and is a non-disruptive solution; service definitions do not have to be altered prior to enactment.

Efficiently executing large-scale, data-intensive workflows such as Montage must take into account the volume and pattern of communication. When orchestrating data-centric workflows, centralised servers common to standard workflow systems can become a bottleneck to performance. However, standards-based workflow systems that rely on centralisation, e.g., Web service based frameworks, have many other benefits such as a wide user base and sustained support.